[kwlug-disc] High Availability

Mon Feb 24 12:51:10 EST 2014

Back probably around 2010 or so I think that I failed resources from a 
slave back to a master using heartbeat. However, I don't recall the 
version of heartbeat or what the command was ... But I think it is 
possible depending on the setup.

I might have old notes stored somewhere.

On 2014-02-21 10:39, Joe Wennechuk wrote:
> So I was thinking there would be a command that I could issue to 
> fallback
>
> or some script that would handle the failback The master node is back
> up, but the service is still pointing to the slave.
>
> ----------------------------------------
>> CC: kwlug-disc at kwlug.org
>> From: jpoole at digitaljedi.ca
>> Date: Thu, 20 Feb 2014 16:07:02 -0500
>> To: kwlug-disc at kwlug.org
>> Subject: Re: [kwlug-disc] High Availability
>>
>> From what I remember with heartbeat setups (a long long time ago) it 
>> was always a manual failback as to control split brain possibilities.
>>
>> Sent from my iPhone
>>
>>> On Feb 20, 2014, at 3:17 PM, Digimer <lists at alteeve.ca> wrote:
>>>
>>> Ok, so I asked on IRC for you. The person I was thinking of is only 
>>> familiar with heartbeat v3. He thinks though that older heartbeat 
>>> doesn't have a "fail-back" mode, it only moves in response to a node 
>>> failure/withdraw.
>>>
>>> Of course, neither of us are sure on this.
>>>
>>> digimer
>>>
>>>> On 20/02/14 12:02 PM, Digimer wrote:
>>>> I'm afraid that I last used heartbeat in 2007 or 2008. It's long
>>>> deprecated and the use base is shrinking fast.
>>>>
>>>> I would suggest you jump on freenode.net, join #linux-ha and ask 
>>>> for
>>>> help there. Also, you could try #drbd as LinBit, the company 
>>>> behind
>>>> DRBD, maintains the heartbeat code base. There is one user there I 
>>>> know
>>>> who is still familiar with heartbeat who might be able to help.
>>>>
>>>> In the meantime, I _strongly_ recommend starting the process of
>>>> migrating the client to corosync + pacemaker. That's the new stack 
>>>> that
>>>> everything is moving towards (even Red Hat is going there in RHEL 
>>>> 7 and
>>>> dropping their own cman + rgmanager).
>>>>
>>>> digimer
>>>>
>>>>> On 20/02/14 11:15 AM, Joe Wennechuk wrote:
>>>>> Ha resources on the slave is ..
>>>>> OSCAR-SLAVE 10.240.7.190 tomcat6 smartlabs inadyn
>>>>> MailTo::support at prylynx.com
>>>>>
>>>>> and on the master is ...
>>>>> OSCAR-MASTER 10.240.7.190 tomcat6 smartlabs inadyn
>>>>> MailTo::support at prylynx.com
>>>>>
>>>>>
>>>>> ----------------------------------------
>>>>>> From: youcanreachmehere at hotmail.com
>>>>>> To: kwlug-disc at kwlug.org
>>>>>> Subject: RE: [kwlug-disc] High Availability
>>>>>> Date: Thu, 20 Feb 2014 11:11:57 -0500
>>>>>>
>>>>>> I have just acquired an already set up customer. It appears that 
>>>>>> it
>>>>>> is using heartbeat.
>>>>>>
>>>>>> haresources on both machines is ....
>>>>>> OSCAR-MASTER 10.240.7.190 tomcat6 smartlabs inadyn
>>>>>> MailTo::support at prylynx.com
>>>>>>
>>>>>> and the ha.cf is ....
>>>>>> logfile /var/log/ha-log
>>>>>> logfacility local0
>>>>>> keepalive 2
>>>>>> udpport 694
>>>>>> bcast eth0 # Linux
>>>>>> auto_failback off
>>>>>> node OSCAR-MASTER
>>>>>> node OSCAR-SLAVE
>>>>>>
>>>>>> OSCAR-MASTER IP is 10.240.7.191
>>>>>> OSCAR-SLAVE IP is 10.240.7.192
>>>>>>
>>>>>> The slave has the ball because if I ifconfig on the slave I get 
>>>>>> the
>>>>>> following that shows the slave is now IT.
>>>>>>
>>>>>> eth0 Link encap:Ethernet HWaddr 10:9a:dd:5b:a3:81
>>>>>> inet addr:10.240.7.192 Bcast:10.240.7.255 Mask:255.255.255.0
>>>>>> inet6 addr: fe80::129a:ddff:fe5b:a381/64 Scope:Link
>>>>>> UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
>>>>>> RX packets:25668625 errors:0 dropped:0 overruns:0 frame:0
>>>>>> TX packets:24503891 errors:0 dropped:0 overruns:0 carrier:0
>>>>>> collisions:0 txqueuelen:1000
>>>>>> RX bytes:865187034 (865.1 MB) TX bytes:1338963431 (1.3 GB)
>>>>>> Interrupt:21
>>>>>>
>>>>>> eth0:0 Link encap:Ethernet HWaddr 10:9a:dd:5b:a3:81
>>>>>> inet addr:10.240.7.190 Bcast:10.240.7.255 Mask:255.255.255.0
>>>>>> UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
>>>>>> Interrupt:21
>>>>>>
>>>>>>
>>>>>>
>>>>>> ----------------------------------------
>>>>>>> Date: Thu, 20 Feb 2014 10:50:11 -0500
>>>>>>> From: lists at alteeve.ca
>>>>>>> To: kwlug-disc at kwlug.org
>>>>>>> Subject: Re: [kwlug-disc] High Availability
>>>>>>>
>>>>>>>> On 20/02/14 10:32 AM, Joe Wennechuk wrote:
>>>>>>>> I have a High Availability cluster, and the slave has taken 
>>>>>>>> over.
>>>>>>>> The master is back up, but failback is not on. Is there any 
>>>>>>>> command
>>>>>>>> to force the failback to the master?
>>>>>>>
>>>>>>> What cluster stack? How is it configured? What is the current 
>>>>>>> reported
>>>>>>> state?
>>>>>>>
>>>>>>> --
>>>>>>> Digimer
>>>>>>> Papers and Projects: https://alteeve.ca/w/
>>>>>>> What if the cure for cancer is trapped in the mind of a person 
>>>>>>> without
>>>>>>> access to education?
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> kwlug-disc mailing list
>>>>>>> kwlug-disc at kwlug.org
>>>>>>> http://kwlug.org/mailman/listinfo/kwlug-disc_kwlug.org
>>>>>
>>>>> _______________________________________________
>>>>> kwlug-disc mailing list
>>>>> kwlug-disc at kwlug.org
>>>>> http://kwlug.org/mailman/listinfo/kwlug-disc_kwlug.org
>>>
>>>
>>> --
>>> Digimer
>>> Papers and Projects: https://alteeve.ca/w/
>>> What if the cure for cancer is trapped in the mind of a person 
>>> without access to education?
>>>
>>>
>>> _______________________________________________
>>> kwlug-disc mailing list
>>> kwlug-disc at kwlug.org
>>> http://kwlug.org/mailman/listinfo/kwlug-disc_kwlug.org
>>
>>
>> _______________________________________________
>> kwlug-disc mailing list
>> kwlug-disc at kwlug.org
>> http://kwlug.org/mailman/listinfo/kwlug-disc_kwlug.org
>
> _______________________________________________
> kwlug-disc mailing list
> kwlug-disc at kwlug.org
> http://kwlug.org/mailman/listinfo/kwlug-disc_kwlug.org

-- 
Andrew Mercer
Kitchener, ON

www.andrewmercer.net